Based on the problems with non sinusoidal converters, i think the whole issue is lays within the PFC circuit.
I think it should not that hard to figure out, what kind of PFC controller chip they used and to compare the circuit with the one from the chip`s data sheet.
I think that there is a design flaw, which stresses some silicon chip repeatedly until it goes boom (probably the PFC inverters switcher).
May be they used a bad toroid core or there is one of those issues you first see in the field.
Reading the post above
"being supplied via an UPS" that puts a slightly different light on it.
Oh ensure brain in gear before typing so I was thinking old school on the bridge thing
I pulled mine out and had a look (didn't break the seal)
From what I can tell (with mag glass sunlight eyestrain etc)
The supposed bridge that I was talking about, It actually looks a little shorter but th pcb has very near it BRD1
The number as best I can get off that semiconductor is its an ST component W12NK80Z which makes it a MOSFET
from the search I did
(really hard to tell NK on it)
Near this marked as U1 an UC3842BN with a couple of opto couplers near by EL 817
(find pdfs on UC3842 here)
(with that number and no expert here but I would have thought most designs would have been tried and trued by now?)
The vertical PCB on the mains side of things I just cant see so no idea what chips are on that. (think this is A-PFC controller pcb?)
There is on the secondary side an IC numbered J05947A1F (PS223)
"logo" square type box with a box (solid) diagonal inside but over lapse edge
no idea what that one is
Anyway with out stripping it down thats the best I can come up with on whats in the box.
cheers
You step into the Road, and if you don't keep your feet, there is no knowing where you may be swept off to." Bilbo Baggins ...
I pulled the cTc caps out. Notice that the 3300uf 10v are the same size as the 3300uf 16v.
Replaced all except the 16v with Rubycons. Xicon for the 3300uf 16v. Don't know if that is a good brand or not. I may replace them if I get better quality caps for that size.
I did look in the Bios and compared voltage readings before and after cap replacement and readings are exactly the same. I'm typing this on the system that has this power supply.
I did notice the slight buzzing that has been talked about at boot. It starts out buzzing for about a half second and then its gone. It is not loud and I can't hear it if I'm 2 feet away from the power supply.
I put a temp probe that hooks up to the motherboard. It is touching a heatsink fin and reading 42c/107f. Not very hot.
Well hopefully this power supply will last. I'm not using the full capacity either which may be prolonging its life. Only 215 Watts reading on my Seasonic Power Angel. I'm afraid to use it when I upgrade the system knowing that my wattage usage will be higher.
Well to me they look like Ferrite beads but not the guru here
(RF suppression? or maybe more some aspect along these lines here )
They look like they are fixed to the legs,
maybe they just come as part of the device...
don't know
The device is marked as Q5
(the one that did the July 4th impression)
Q4 is on the opposite side and is the same type
Q2 and Q3 (Q2,4 mains side Q3,5 inside.)
Just cant get the type numbers off Q3 or Q5
(3 devices in all on each side of the first heatsink from what I can tell)
Q2~Q5 have these on the outside legs.
photo of Q4
Just having a look at the infiniti photos (above review)
The mains side is similar but different
I think the A-PFC (Pre Reg) PCB is the same to look at
(not to say there isn't component changes)
There is defiantly changes when you start comparing the 2 units
from the photos
(wish I had an Infiniti to pull apart oh well)
Cheers
Attached Files
You step into the Road, and if you don't keep your feet, there is no knowing where you may be swept off to." Bilbo Baggins ...
Xicon no idea on those caps and ruby's I suppose as you know, it depends on what series
They do look like the typical anorexic caps I'll have to check the diameter.
(and spacing)
Yours from what I can tell look ok and my guess is under light loads they will last a while...but they are not the cream of the crop and for a PSU that was expensive as they were So "not impressed Jan!" (err a tv advert we have)
Ideally you probably wouldn't get a Voltage change and yeah a real Multimeter might show a slight difference.
The real test is in heat and high current demands over time I guess.
Yes that was my worry,
(and the reason I bought the liberty was with upgrade in mind)
Since finding out about the Flash Bang Factor
I decided not worth the Risk and got a Corsair HX620W to run this box on
(not that it pulls all that much I suppose just wasn't worth the risk if I did start putting in stuff that did chew juice)
(cheaper then the liberty too )
So the liberty will run the old box somewhat stripped down now.
My older enermax is still going,
I think they stuffed up with the Liberty range and it is going to cost them.
It would be nice to have a resolve to this issue.
Especially if it is a common one and possible design flaw.
For those liberty owners thats suffered from it
A repair regardless of warranty would go a long way in limiting the damage from this issue.
I guess in the high end game a lot of sales are word of mouth from other users recommendations and for the Liberty series this don't look good.
(somewhat of a mute point as I think the Liberty range is history but still)
Cheers
Oh yeah think that was suppose to be UC3842BN on the photo
Update:
My Liberty is still running with no problems yet. My temps are higher than I posted earlier. My temps go up to 50c having my probe touch the smaller heatsink. 55 is the highest I've seen. It seems the pdfs you posted starfury1 show that temps for the components go up to 150c. I originally thought that high temps maybe killing the parts.
The fan speed remains the same even when the powersupply load was 255 watts when gaming. I do not know if the fan controller will ramp up the RPM's if the heat or watts go up. Currently my motherboard shows 912 to 924 RPM's.
What I think we all have learned from all this is that we cannot rely on this power supply at its full capacity and I will replace it and put it in a low wattage system when I upgrade my motherboard in the future. For now it is doing its job.
as to the fan...humm well I think? they are suppose to gear up when the temp raises but when at what point I don't know.
there is defiantly a question mark on them that is for sure but to quote Super Nade from his blog, part of it anyway see his link for the text in total, I should add... not in reference to the liberty but still a true statement regasrdless
Hope you dont mind Super Nade ?????
It only takes one batch of defective units to create an avalanche of bad press..
Very true
Still it is of concern
I do have more photos now and just haven't posted them as thread seems to have dropped off a bit.
That Fet is by the looks of it, is controlled from the 3842 chip
I did notice some of the solder joints (to me) looked a bit iffy
(I should have reworked them But think its RoHs tin solder I'll have to confirm this)
don't know about the temps but if you wattage is correct thats a little over a third of what it should be capable of.
I should think it should do that no worries but I think Gal mention one dieing under a light load so that to me says something else is happing
I guess its the unknown factor here
Do you have a psu that could die at anytime?
which is why I decided not to chance it.
Even though it might only be a very small chance.
Be nice if someone that knows their stuff got hold of blown one for an analysis
but anyway I'll use mine in the old box were hopefully it will live along and happy life.
If anyone really wants to see some more photos post and I'll put them up.
I should think it should do that no worries but I think Gal mention one dieing under a light load so that to me says something else is happing
Keep in mind that besides a lot of them failing from some flaw (presuming there is a flaw), that doesn't preclude any other failure mode a certain % would be expected to see if there were no flaw(s) present. It could be a lot of different things, even a bad reel of FETs but then once they use up that reel and pop another on, no more problems. If I had one I think I would just replace that FET with the studliest part I could find, put an inrush current limiter in front of it, then use it as if nothing happened.
True!
Its only being surmised that its the exact same type of failure,
from what I gather.. but yes it could be different...
thats why it would be nice to know.
There is from what I gather no real hard core facts on this only they went "flash Bang"
Plywood99's is the first internal photos I have seen of it.
I guess most were RMAed...its an expensive supply to just write of if its under warranty
My other thought, is how many where being used on an UPS when they Flash Banged...its suppose to be used on a Sinewave inverter type ups
Its just another possibility really.
I am hoping it is there case, that it was just a few but we got no way of telling
I did read somewhere of other liberties dieing in the lower wattages
so humm more food for thought I suppose
As you say a bit of pre-emptive work might solve the possible potential flaw.
Its the not knowing exactly what it is thats the hard part.
The Infiniti I have not heard of problems
Although some different they look somewhat same too
I guess a possible lemon can happen to any company,
Liberties my have been Enermax's
(or possible Lemon really)
The original post Thread starter, was nothing to do with this issue it was popped CTC caps.
That can happen to any supply, give hard work heat and possibly not great caps, even great ones are going to die under extreme conditions
So the conditions of use may well have contributed to the "flash bang" issue so that needs to be taken into account
Anyway for those that might make a bit more sense out of it here are a few more photos
Cheers
Attached Files
You step into the Road, and if you don't keep your feet, there is no knowing where you may be swept off to." Bilbo Baggins ...
Is the one that's vented installed with the wrong polarity?
Mann-Made Global Warming. - We should be more concerned about the Intellectual Climate.
-
Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind. - Dr Seuss - You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
-
Well it finally happened. The Enermax died 2 weeks ago and I'm finally posting this. The system that the powersupply is connected to is only using about 230 watts of power.
So here is what happened. I got up for breakfast and while eating away the lights flicker slightly and at the same time the UPS start beeping like I've never heard it before. So I run to find that the computer is off and the UPS overload LED is on. I have this pc on all the time running Folding@home.
So I shut it down and pull the powersupply out. Already knowing what to expect. It was only warm to the touch. No visual damage so I plug in a powersupply tester and plug directly to the AC outlet. I flip the switch on the Enermax and well...
You know what happens. A pop and bright flash, then some sparks exit the exhaust area. A nice greenish light occurs for about 1 sec. Then nothing but some smoke and all is quiet. I then unplugged it and proceeded to open it up.
Here is what I found.
And so I ordered a new Corsair HX620W. The computer is up and running and I'm very glad the Enermax caused no damage other than to itself. I'm glad the Conext UPS shutdown before the powersupply blew up. I do not know if having a UPS connected would have caused the problem but a powersupply company should know that users could be using one and should take that into account in there design.
Have a Merry Christmas everyone. Or happy Kwanzaa, Hanukkah.
I recapped this unit and it has since been running fine, but after reading about yet another flashbang......
Time to quit using this thing.
It definitely does have a flaw in the PFC circuitry. The flashbang issue has all been in the same components. Too bad Enermax is refusing to liven up that such a problem really does exist.
UPS start beeping like I've never heard it before.
I am really now wondering just how many of these go boom on an UPS
Gal
13 months after the RMA a second Liberty flash-bangs on Minerva.
She is not a n00b as I noted earlyer in this thread.
Yep I know shes not, you told me and from memory the first was on a ship.
(i'll read the post later)
Tell here sorry to hear that too
Again this brings me back to the fact they say only use with sine wave inverter
(or whatever, UPS)
Spyke you were the thread starter and yeah I think you give it a somewhat brutal workout
So maybe it would be best in your case...stay away from enermax period in your case I think
But I'll ask the question .."Do you use an UPS ?"
I dont on any of mine
The Enermax has been put out to pasture more or less so it does not get the work out it should have been getting.
Comment